Published Paper Details:

The method of identifying the handwritten text using machine interface is known as handwritten character recognition. Handwritten characters present challenges because they differ from one writer to the other; even though the same author writes the same character, the form, size, and location of the character vary. Handwritten Character Recognition has been one of the challenging areas in the field of image processing. It has numerous applications which include, bank cheques and conversion of any handwritten document into digital text form. There is a prominent need for storing information to desktop storage from the information available in handwritten images to later reprocess this information utilizing computers. But to re-process this information, it would be very difficult to read the other information from these image files. Therefore, a technique to automatically resolve and store information, in particular text, from image files is needed. This paper aims to classify an individual handwritten word so that handwritten text for English alphabets can be translated to a typed or editable form. We used the Convolutional Neural Networks to accomplish this task. We used this architecture to train a model that can accurately predict words. Trends suggest that the market value for such applications is going to increase exponentially in the future.